A computational tool for the probabilistic finite element analysis of an uncemented total hip replacement considering variability in bone–implant version angle

Abstract

In the present study, a probabilistic finite element tool was implemented to assess an uncemented total hip replacement including variability in bone-implant version angle. The Monte Carlo method was used with two different performance indicators: the bone maximum nodal von-Mises elastic strain and the bone volume (BV) percentage exceeding specified strain limits. Implant version, bone stiffness and load magnitude were the most sensitive parameters. The results were more consistent using percentage BV under specified limit strains as the performance indicator, even for a low number of simulations. The reliability of the computational tool was demonstrated through a comparison with previous studies, and the consistency of the results for all strain limits investigated.

Probabilistic finite element analysis of the uncemented hip replacement—effect of femur characteristics and implant design geometry

“…In addition, further resources are needed if bone geometry and material uncertainties are to be considered, such as a substantial sample of bones, representative of the population. Implant material variability was found to be minimally sensitive in a previous study (Dopico-Gonzalez et al, 2009a) and hence it was neglected for the present study; some studies have suggested the use of strainbased failure criteria as fracture predictors (Wong et al, 2001;Kopperdahl, 1999;Schileo et al, 2007;Morgan and Keaveny, 2001;Bayraktar et al, 2004), similar to that adopted in the present work. It is assumed that the femur will break when a certain amount of bone suffers strains over a limit close to the yield strain, believed to be approximately 0.78%70.06 (Kopperdahl, 1999).…”

“…The reference geometry was determined for the femur using MATLAB modules, as in a previous study (Dopico-Gonzalez et al, 2009a).…”

“…Some researchers have developed computational tools that enable analysis of multiple scenarios using a set of random variables simultaneously and efficiently (Browne and Langley, 1999;Nicolella et al, 2001;Bah and Browne, 2003;Perez et al, 2006;Easley et al, 2007;Mehrez, 2007). In probabilistic studies on THR models, it has been shown that there are many important random variables, such as bone geometry, implant design geometry, cement geometry, material properties, magnitude and direction of applied loads, and bone-implant relative position that can affect performance (Browne and Langley, 1999;Nicolella et al, 2001;Bah and Browne, 2003;Perez et al, 2006;Easley et al, 2007;Mehrez, 2007;Dopico-Gonzalez et al, 2009a;Dopico-Gonzalez et al, 2009b;Halpern and Tanner, 1979;Heller et al, 2001;Herrlin et al, 1988;Laz et al, 2007a;Nicolella et al, 2006;Nishii et al, 2004;Reikeras et al, 1982;Speirs et al, 2007;Yang et al, 1984). This suggests that it is not appropriate to extrapolate the results obtained from simulation studies of one model to the whole population of femurs and implant designs.…”

“…There were two objectives of the present study: (i) To further develop a computational tool (Dopico-Gonzalez et al, 2009a;Dopico-Gonzalez et al, 2009b) for the probabilistic analysis of the uncemented hip replacement by constructing a more realistic finite element (FE) model and by including 6 degrees of freedom of bone-implant relative position parameters in the set of random input variables and (ii) to assess the effect of femur characteristics and implant design geometry on the results. To this end, three femurs and two implant designs, the Proxima short stem (DePuy, Warsaw, USA) and the IPS stem (DePuy, Leeds, United Kingdom) were investigated using Monte Carlo-based simulations.…”

A review of probabilistic analysis in orthopaedic biomechanics

“…A Monte Carlo analysis was performed on a hip stem construct to evaluate the likelihood of instability or micromotion based on uncertainty in the bone material properties, body weight loading, implant size, and the region of contact [25]. Considering uncertainty in bone property, loading, and also component alignment in a hip stem construct, Dopico-Gonzalez et al applied probabilistic methods to find that the volume of strained bone was sensitive to the implant version angle, bone modulus, and applied load; good agreement was demonstrated for Monte Carlo and Latin hypercube sampling [26,27]. In a study evaluating the response of three femurs to two types of implants, one with a stem and one without a stem, Dopico-Gonzalez et al showed that micromotion in both implants was most sensitive to implant positioning parameters (Fig.…”